Efficient Incentive Scheme forWireless Random Channel Access with Selfish Users

Abstract

In the decentralized networks, interactions among selfish users sharing a common transmission channel can be modeled as a non cooperative game using game theory. When selfish users choose their transmission probabilities independently without any coordination mechanism, Nash equilibria leads to a suboptimal use of the channel resource and to the degradation of the performance of MAC protocols. In this paper we transform the non cooperative game in a general Stackelebrg game when multiple leaders and followers coexist and competitively maximize their own function utilities. Under a simple multiple power levels scheme, the new Stackleberg equilibrium can overcome the deficiency of the Nash equilibrium as well as the limited efficiency of the Sackelberg contention game when the users transmit using a single power level. The power control is defined in such way that the leaders choose the lowest power to transmit their packets among N available levels whereas the followers retransmit at the random power levels picked from $N-1$ higher distinct. Furthermore, we investigate the impact of the implementation of our scheme on the equilibrium and discuss the optimal partition of leaders/followers to achieve better performance. We further show that the hierarchical system is performing better as the number of leaders is small.